Rapid assessment of breast tumor margins using deep ultraviolet fluorescence scanning microscopy

Deep learning classification of deep ultraviolet fluorescence images toward intra-operative margin assessment in breast cancer

Background

Breast-conserving surgery is aimed at removing all cancerous cells while minimizing the loss of healthy tissue. To ensure a balance between complete resection of cancer and preservation of healthy tissue, it is necessary to assess themargins of the removed specimen during the operation. Deep ultraviolet (DUV) fluorescence scanning microscopy provides rapid whole-surface imaging (WSI) of resected tissues with significant contrast between malignant and normal/benign tissue. Intra-operative margin assessment with DUV images would benefit from an automated breast cancer classification method.

Methods

Deep learning has shown promising results in breast cancer classification, but the limited DUV image dataset presents the challenge of overfitting to train a robust network. To overcome this challenge, the DUV-WSI images are split into small patches, and features are extracted using a pre-trained convolutional neural network-afterward, a gradient-boosting tree trains on these features for patch-level classification. An ensemble learning approach merges patch-level classification results and regional importance to determine the margin status. An explainable artificial intelligence method calculates the regional importance values.

Results

The proposed method's ability to determine the DUV WSI was high with 95% accuracy. The 100% sensitivity shows that the method can detect malignant cases efficiently. The method could also accurately localize areas that contain malignant or normal/benign tissue.

Conclusion

The proposed method outperforms the standard deep learning classification methods on the DUV breast surgical samples. The results suggest that it can be used to improve classification performance and identify cancerous regions more effectively.

Rapid digital pathology of H&E-stained fresh human brain specimens as an alternative to frozen biopsy

BACKGROUND

Hematoxylin and Eosin (H&E)-based frozen section (FS) pathology is presently the global standard for intraoperative tumor assessment (ITA). Preparation of frozen section is labor intensive, which might consume up-to 30 minutes, and is susceptible to freezing artifacts. An FS-alternative technique is thus necessary, which is sectioning-free, artifact-free, fast, accurate, and reliably deployable without machine learning and/or additional interpretation training.

METHODS

We develop a training-free true-H&E Rapid Fresh digital-Pathology (the-RFP) technique which is 4 times faster than the conventional preparation of frozen sections. The-RFP is assisted by a mesoscale Nonlinear Optical Gigascope (mNLOG) platform with a streamlined rapid artifact-compensated 2D large-field mosaic-stitching (rac2D-LMS) approach. A sub-6-minute True-H&E Rapid whole-mount-Soft-Tissue Staining (the-RSTS) protocol is introduced for soft/frangible fresh brain specimens. The mNLOG platform utilizes third harmonic generation (THG) and two-photon excitation fluorescence (TPEF) signals from H and E dyes, respectively, to yield the-RFP images.

RESULTS

We demonstrate the-RFP technique on fresh excised human brain specimens. The-RFP enables optically-sectioned high-resolution 2D scanning and digital display of a 1 cm² area in <120 seconds with 3.6 Gigapixels at a sustained effective throughput of >700 M bits/sec, with zero post-acquisition data/image processing. Training-free blind tests considering 50 normal and tumor-specific brain specimens obtained from 8 participants reveal 100% match to the respective formalin-fixed paraffin-embedded (FFPE)-biopsy outcomes.

CONCLUSIONS

We provide a digital ITA solution: the-RFP, which is potentially a fast and reliable alternative to FS-pathology. With H&E-compatibility, the-RFP eliminates color- and morphology-specific additional interpretation training for a pathologist, and the-RFP-assessed specimen can reliably undergo FFPE-biopsy confirmation.

Analysis of Deep Ultraviolet Fluorescence Images for Intraoperative Breast Tumor Margin Assessment

Positive margin status after breast-conserving surgery (BCS) is a predictor of higher rates of local recurrence. Intraoperative margin assessment aims to achieve negative surgical margin status at the first operation, thus reducing the re-excision rates that are usually associated with potential surgical complications, increased medical costs, and mental pressure on patients. Microscopy with ultraviolet surface excitation (MUSE) can rapidly image tissue surfaces with subcellular resolution and sharp contrasts by utilizing the nature of the thin optical sectioning thickness of deep ultraviolet light. We have previously imaged 66 fresh human breast specimens that were topically stained with propidium iodide and eosin Y using a customized MUSE system. To achieve objective and automated assessment of MUSE images, a machine learning model is developed for binary (tumor vs. normal) classification of obtained MUSE images. Features extracted by texture analysis and pre-trained convolutional neural networks (CNN) have been investigated for sample descriptions. A sensitivity, specificity, and accuracy better than 90% have been achieved for detecting tumorous specimens. The result suggests the potential of MUSE with machine learning being utilized for intraoperative margin assessment during BCS.

Automated assessment of breast margins in deep ultraviolet fluorescence images using texture analysis

Microscopy with ultraviolet surface excitation (MUSE) is increasingly studied for intraoperative assessment of tumor margins during breast-conserving surgery to reduce the re-excision rate. Here we report a two-step classification approach using texture analysis of MUSE images to automate the margin detection. A study dataset consisting of MUSE images from 66 human breast tissues was constructed for model training and validation. Features extracted using six texture analysis methods were investigated for tissue characterization, and a support vector machine was trained for binary classification of image patches within a full image based on selected feature subsets. A weighted majority voting strategy classified a sample as tumor or normal. Using the eight most predictive features ranked by the maximum relevance minimum redundancy and Laplacian scores methods has achieved a sample classification accuracy of 92.4% and 93.0%, respectively. Local binary pattern alone has achieved an accuracy of 90.3%.

Deep Learning for Breast Cancer Classification of Deep Ultraviolet Fluorescence Images toward Intra-Operative Margin Assessment

Breast conserving surgery aims at the complete removal of malignant lesions while minimizing healthy tissue loss. To ensure the balance between complete resection of the cancer and conservation of healthy tissue, intra-operative margin assessment is necessary. Deep ultraviolet (DUV) fluorescence scanning microscope provides fast whole-surface-imaging (WSI) of excised tissue with contrast between malignant and normal tissues. Then, an automated breast cancer classification method on DUV images is required for intra-operative margin assessment. Deep learning shows the promising results in breast cancer classification, but limited DUV image dataset poses overfitting challenge to train the robust network. To tackle this challenge, we partition the DUV WSI image into small patches and extract pathological features for each patch from a pre-trained network using a transfer learning approach. We feed pathological features into a decision-tree-based classifier and fuse patch-level classification results based on regional importance to determine malignant or benign WSI. Experimental results on 60 DUV images show that our proposed method outperforms the standard deep learning classification in terms of improving the classification performance and identifying cancerous regions.

Radiofrequency spectroscopy with intraoperative pathological assessment for breast carcinoma: synergistic or redundant?

BACKGROUND

Published MarginProbe (Dune Medical Devices Ltd., Israel) data reports ≥50% reduction in positive lumpectomy margins. We sought to determine whether adjunctive use of MarginProbe would provide value over intraoperative pathologic assessment alone.

METHODS

This is a retrospective chart review of 86 consecutive lumpectomies with MarginProbe from December 2018 to November 2019. Margins were considered positive using 'no ink on tumor' guideline for invasive cancer, and 2 mm or greater for ductal carcinoma in-situ. Significance was measured using Fisher's exact test.

RESULTS

Seventy-six patients yielded 86 lumpectomies for inclusion. Mean age was 69.8 and mean tumor size was 1.09 cm. Sixty-eight invasive cancers were assessed using adjunct MarginProbe and gross assessment, while 18 ductal carcinoma in-situ cases utilized MarginProbe only. Among all cases, gross assessment alone reduced positive margins(29.2% relative reduction, p = 0.28). Utilizing both modalities, positive margins decreased from 27.9% to 9.3% (66.7% relative reduction, p < 0.01) representing a 46.9% relative reduction versus gross assessment alone. After gross assessment and MarginProbe evaluation, additional excised volume averaged 2.9 cc.

CONCLUSIONS

Synergistic use of MarginProbe and gross assessment reduces positive margins during breast conserving surgery. Surgeons can weigh its cost against it benefit with the succinct analysis we provide.

The Critical Role of Breast Specimen Gross Evaluation for Optimal Personalized Cancer Care

Gross examination is the foundation for the pathologic evaluation of all surgical specimens. The rapid identification of cancers is essential for intraoperative assessment and preservation of biomolecules for molecular assays. Key components of the gross examination include the accurate identification of the lesions of interest, correlation with clinical and radiologic findings, assessment of lesion number and size, relationship to surgical margins, documenting the extent of disease spread to the skin and chest wall, and the identification of axillary lymph nodes. Although the importance of gross evaluation is undeniable, current challenges include the difficulty of teaching grossing well and its possible perceived undervaluation compared with microscopic and molecular studies. In the future, new rapid imaging techniques without the need for tissue processing may provide an ideal melding of gross and microscopic pathologic evaluation.

Improved microscopy with ultraviolet surface excitation (MUSE) using high-index immersion illumination

Microscopy with ultraviolet surface excitation (MUSE) typically has an optical sectioning thickness significantly larger than standard physical sectioning thickness, resulting in increased background fluorescence and higher feature density compared to formalin-fixed, paraffin-embedded physical sections. We demonstrate that high-index immersion with angled illumination significantly reduces optical sectioning thickness through increased angle of refraction of excitation light at the tissue interface. We present a novel objective dipping cap and waveguide-based MUSE illuminator design with high-index immersion and quantify the improvement in optical sectioning thickness, demonstrating an e^-1 section thickness reduction to 6.67 µm in tissue. Simultaneously, the waveguide illuminator can be combined with high or low magnification objectives, and we demonstrate a 6 mm² field of view, wider than a conventional 10x pathology objective. Finally, we show that resolution and contrast can be further improved using deconvolution and focal stacking, enabling imaging that is robust to irregular surface profiles on surgical specimens.

Optical scatter imaging of resected breast tumor structures matches the patterns of micro-computed tomography

In patients undergoing breast-conserving surgery (BCS), the rate of re-excision procedures to remove residual tumor left behind after initial resection can be high. Projection radiography, and recently, volumetric x-ray imaging are used to assess margin adequacy, but x-ray imaging lacks contrast between healthy, abnormal benign, and malignant fibrous tissues important for surgical decision making. The purpose of this study was to compare micro-CT and optical scatter imagery of surgical breast specimens and to demonstrate enhanced contrast-to intra-tumoral morphologies and tumor boundary features revealed by optical scatter imaging. A total of 57 breast tumor slices from 57 patients were imagedex vivoby spatially co-registered micro-CT and optical scatter scanning. Optical scatter exhibited greater similarity with micro-CT in 89% (51/57) of specimens versus diffuse white light (DWL) luminance using mutual information (mean ± standard deviation of 0.48 ± 0.21 versus 0.24 ± 0.12;p < 0.001) and in 81% (46/57) of specimens using the Sørensen-Dice coefficient (0.48 ± 0.21 versus 0.33 ± 0.18;p < 0.001). The coefficient of variation (CV) quantified the feature content in each image. Optical scatter exhibited the highest CV in every specimen (optical scatter: 0.70 ± 0.17; diffuse luminance: 0.24 ± 01; micro-CT: 0.15 ± 0.03 for micro-CT;p < 0.001). Optical scatter also exhibited the highest contrast ratios across representative tumor boundaries with adjacent healthy/benign fibrous tissues (1.5-3.7 for optical scatter; 1.0-1.1 for diffuse luminance; 1.0-1.1 for micro-CT). The two main findings from this study were: first, optical scatter contrast was in general similar to the radiological view of the tissue relative to DWL imaging; and second, optical scatter revealed additional features associated with fibrous tissue structures of similar radiodensity that may be relevant to diagnosis. The value of micro-CT lies in its rapid three-dimensional scanning of specimen morphology, and combined with optical scatter imaging with sensitivity to fibrous surface tissues, may be an attractive solution for margin assessment during BCS.